Models

phi35-sap-ax-merged

gemma-2-9b-it-sae-scoped-coding

llama2_7b_chat-WaRP-gsm8k-FT-lr3e-5_ssft_5e-5

llama-3_1-8b-simnpo-gentle-bm25-6t

CRRL_distill_1.5B_GRESO_step_90

Forgotten-Abomination-24B-V3.0

Qwen2.5-Coder-LEAK-MCEVALHARD-1.5B-Base-1

qwen2.5-3B-sql-mgpu-bi-ft

qwen3-8b-simnpo-gentle-bm25-6t

smartclaims-grpo-unk10

llama-2-13b-chat-hf-gsm8k-rsn-tuned-lr5e-5

Qwen3-8B_julia_codeforces_with_thinksft_16bit_vllm

Qwen3_Without_COT

Mlem-4B-RL-Thinking-Seed1

qwen3-8b-undial-baseline-target-100

llama-3_1-8b-simnpo-gentle-baseline

gemma-2-9b-it-lr3e-5-safeinstr-0.1

qwen3-4b-35b-rk-new_solver_aux_v4

Latent-SFT-Llama3.2-Instruct-1B-COT-SFT

filing-sense-grpo-qwen2.5-3b

sn38-v11-2

Qwen3-0.6B-Base-CPT-Math

affine-5F4JyqstSdvMfZcRuFvyAGPer25Cu1PmNd3snnHfaA7gxguZ

Magidonia-24B-v4.3-heretic-v1.2

llama-3_1-8b-simnpo-gentle-baseline-target-100

opsd_2b_lora_2k

Magro-7b-v1.1

OceanGPT-basic-7B-v0.3

Gemma-3-4B-IT-GA-SynthDolly-1A-E3

Qwen3-8B-Base-baseline-ghpo

qwen2.5-7b-cabs-v0.4

llama3.1_8b_base-SSFT-start-WaRP-original-space-gsm8k-FT-lr3e-5

Qwen2.5-Coder-LEAK-MCEVALHARD-1.5B-Base-9

Qwen3-1.7B-Base-dapo_filter-prm-eta100-Advorm-stepsplit-none

Hush-Qwen2.5-7B-MST-v1.3

gptlong_continue_nemotron_terminal_step900__Qwen3-32B

tezos100k_continue_top8diverse100k_step3000__Qwen3-32B

g1_top8_85k_gptlong_swegym_32b_step4200__Qwen3-32B

tezos100k_continue_gptlongtezos_step1200__Qwen3-32B

DeepSeek-R1-Distill-Qwen-1.5B-GRPO

Qwen3-4B_CRRL_batch_1024_B200_ds_samplelevelmean_step_110

0416_retrain_merged